diff_of_means ratio_of_sd amplitude_ratio_of_means maximum_error ks_mean_on_coarse_res_with_extremes rainy_hours_ratio_of_means qqplot_mae acf_mae extremogram_mae
nv.cesm2.ssp245 -1.10% 0.830 0.539 0.302 0.587 0.815 0.029 0.152 0.109
lstm.cesm2.ssp585 -1.30% 0.870 0.590 0.348 0.292 0.819 0.031 0.115 0.078
lstm.cesm2.ssp370 -1.32% 0.849 0.591 0.340 0.227 0.798 0.033 0.117 0.069
lstm.cesm2.ssp245 1.45% 0.809 0.581 0.339 0.333 0.832 0.032 0.111 0.058
nv.cesm2.ssp585 -4.32% 0.901 0.551 0.315 0.474 0.808 0.026 0.154 0.124
xgboost.cesm2.ssp370 4.57% 0.878 0.526 0.270 0.387 0.901 0.019 0.121 0.090
nv.cesm2.ssp370 -4.78% 0.894 0.552 0.316 0.378 0.792 0.028 0.147 0.117
cnn.cesm2.ssp245 -5.27% 0.916 0.632 0.358 0.324 0.892 0.028 0.088 0.057
xgboost.cesm2.ssp585 5.57% 0.890 0.533 0.267 0.407 0.937 0.017 0.117 0.094
nv.mri_esm2_0.ssp370 6.06% 0.787 0.511 0.275 0.565 0.874 0.029 0.154 0.106
cnn.mri_esm2_0.ssp370 7.44% 0.824 0.574 0.390 0.314 1.005 0.035 0.082 0.056
xgboost.cesm2.ssp245 7.53% 0.857 0.529 0.277 0.436 0.951 0.018 0.110 0.083
lstm.mri_esm2_0.ssp370 7.76% 0.790 0.562 0.319 0.269 0.916 0.033 0.103 0.048
cnn.cesm2.ssp585 -8.86% 1.011 0.649 0.367 0.279 0.886 0.030 0.091 0.078
cnn.cesm2.ssp370 -8.99% 0.984 0.642 0.361 0.246 0.859 0.030 0.081 0.056
xgboost.mri_esm2_0.ssp370 17.31% 0.784 0.487 0.240 0.461 1.060 0.025 0.108 0.078
cnn.mri_esm2_0.ssp434 18.03% 0.778 0.501 0.412 0.255 1.142 0.049 0.088 0.085
nv.mri_esm2_0.ssp434 18.19% 0.734 0.465 0.276 0.473 0.979 0.036 0.153 0.135
xgboost.ec_earth3.ssp434 -18.59% 0.933 0.618 0.285 0.340 0.740 0.037 0.145 0.082
lstm.mri_esm2_0.ssp434 18.71% 0.733 0.500 0.325 0.255 1.012 0.043 0.111 0.078
nv.mri_esm2_0.ssp245 19.99% 0.709 0.451 0.281 0.569 0.995 0.037 0.171 0.161
cnn.mri_esm2_0.ssp245 20.21% 0.738 0.490 0.394 0.262 1.141 0.052 0.095 0.095
lstm.mri_esm2_0.ssp245 20.64% 0.711 0.491 0.333 0.263 1.026 0.044 0.115 0.081
lstm.ec_earth3.ssp434 -22.31% 0.892 0.703 0.346 0.206 0.669 0.046 0.126 0.049
nv.ec_earth3.ssp434 -24.89% 0.884 0.599 0.295 0.442 0.655 0.053 0.182 0.103
xgboost.mri_esm2_0.ssp434 28.62% 0.728 0.427 0.252 0.409 1.192 0.040 0.109 0.106
cnn.ec_earth3.ssp434 -29.17% 0.992 0.758 0.349 0.245 0.712 0.042 0.112 0.041
xgboost.mri_esm2_0.ssp245 30.08% 0.707 0.416 0.248 0.461 1.204 0.041 0.128 0.127

Time series of the first days

How Often Peaks Hit Hourly

QQ Plot

Distribution of the undownscaled value on days with estimated extremes values.

On the x-axis we have the daily mean (standardized). It says Undownscaled value, but is the daily mean after the downscaling. A good idea is to plot the original undownscaled value.

The purpose of this plot is to illustrate the distribution of P(undownscaled value | we predicted an extreme). This is useful because it reveals how much information we can recover concerning extreme events. If the distribution is skewed to the right, it suggests that we’re predicting extreme values only when extreme values have already occurred. Conversely, if the lower tail of the distribution resembles the reanalysis data, it indicates that we can capture short-duration extremes (e.g., brief periods of heavy rainfall, such as an intense downpour lasting an hour before stopping).

Autocorrelogram

Extremogram